Gang Liao
Research Scientist @ Meta
About
I build the systems that make ML work at scale — from silicon to software. At Meta, my work spans AI accelerator pre-silicon validation (ASIC), ads model-hardware co-design, GPU kernel & compiler optimization, model training & serving, data infrastructure, and agentic AI systems.If you're interested in learning more about my work, feel free to visit my homepage at https://gangliao.me.
United States
Mountain View
Computer Software
Data Management, Vector Database, Model Serving/Inference, Database Systems, Operating Systems, Supercomputing, Application-Specific Integrated Circuits (ASIC), Recommender Systems, Advertising, HTAP, Resource Management, Storage, KV Store, Distributed Training, TensorFlow, Large Language Models (LLM), Parquet, Apache Spark, Multi-tenancy, Rust (Programming Language)
Experience
Research Scientist
Meta
Sunnyvale, CA
Leading KernelEvolve (https://arxiv.org/abs/2512.23236) - Shared Data Foundation for Agentic Systems at Meta - Agentic Kernel Optimization on NVIDA, AMD and MTIA - Continual Test Time Learning to Discover AI Kernels
Research Scientist
TikTok
San Jose, California, United States
Column Storage at Infrastructure System Lab - Build a multi-tenant HTAP system that can support tens of millions of tables with great performance and elasticity. - Lead the next-generation columnar format for Data Lakehouse and Machine Learning. - Bullion: A Column Store for Machine Learning (CIDR'25, https://arxiv.org/abs/2404.08901) - SFVInt: Simple, Fast and Generic Variable-Length Integer Decoding using Bit Manipulation Instructions (DaMoN'24, https://arxiv.org/abs/2403.06898) - Lead the vector database project for in-DB similarity search and retrieval-augmented generation.
Software Engineer Intern, Systems and Infrastructure (PhD)
Meta
Menlo Park, California, United States
Built blob caching for key value separation RocksDB: https://github.com/facebook/rocksdb
Research Intern
Microsoft
Redmond, Washington, United States
Data Systems Group (DMX), MSR https://www.microsoft.com/en-us/research/group/database/people/ Worked with Yinan Li, Badrish Chandramouli, and Donald Kossmann Fast Analytics on Fast Data for HTAP and Data Lake
Research Intern
ByteDance
Bellevue, Washington, United States
Machine Learning System Group
Software Engineer
Sogou
Sogou - Elastic Deep Learning Platform for Speech Recognition
Software Engineer
Baidu USA
Built PaddlePaddle: A Parallel Distributed Deep Learning Framework from Industrial Practice https://github.com/PaddlePaddle/Paddle Top 2 contributors in 2016 and 2017: https://github.com/PaddlePaddle/Paddle/graphs/contributors?from=2016-08-31&to=2017-08-10&type=c
Research Intern
Intel Labs
Hierarchical Matrix Computation on Intel Xeon Phi Accelerator
Education
University of Maryland
Computer Science
Advisor: Daniel Abadi (https://en.wikipedia.org/wiki/Daniel_Abadi) Ph.D. Thesis: The Evolution of Cloud Data Architectures: Storage, Compute, and Migration https://drum.lib.umd.edu/handle/1903/29199
KAUST (King Abdullah University of Science and Technology)
High Performance Computing, Super Computer, GPU
Advisor: David E. Keyes (https://en.wikipedia.org/wiki/David_E._Keyes) https://www.hpcwire.com/35-hpc-legends-david-keyes/
Gang Liao's Contact Information
Phone
Find the Right Leads
Find Verified Contact Data
What LeadContact does well
Find verified emails, phone numbers, and decision-makers with 98% accuracy.
Find Leads
Find the right people by company, role, industry, location, and more.
925M+ professional profiles
Find Emails
Access verified email addresses for your target contacts.
657M+ emails
Find Phone Numbers
Get cross-validated phone data from multiple top sources.
239M+ phone numbers
More Accurate. Lower Cost.
Find contact data in 1 tool with 98% accuracy
LeadContact integrates leading enrichment tools to deliver more accurate contact data—without paying for each one.
Great conversations start with the right contact.
It’s time to find yours.